This document proposes a machine learning approach using the Naive Bayes algorithm to detect distributed denial of service (DDoS) attacks through network intrusion detection. It first discusses the issues with existing intrusion detection systems, including long training times and low accuracy. It then summarizes research on applying various machine learning techniques like neural networks, decision trees, and Naive Bayes to intrusion detection. The proposed system would build a classifier using Naive Bayes, which provides faster training than other methods, to distinguish normal and attack traffic. This approach aims to improve upon the training time and detection accuracy of existing intrusion detection systems.